Novel Algorithm for Volumetric Segmentation of the Paranasal Sinuses on CT Scans Download HRPP SOPs Clinical Trials Find a Researcher NJH ID: #20-05 Background Rhinosinusitis is characterized by severe inflammation of the sinus cavities and is considered the most common chronic illness in the U.S., affecting 1 in 7 Americans with evidence of increasing incidence. Computerized tomography (CT) imaging enables noninvasive assessment of the nasal sinuses. Mucosal thickening visible on CT is considered an objective diagnostic criterion for chronic rhinosinusitis. Traditional evaluation of CT is by visual assessment, with result reported in qualitative, imprecise terms that are not necessarily standardized across readers. There is a recognized unmet need for improved, standardized and objective assessment of sinus opacification, disease severity, and treatment response in chronic rhinosinusitis (CRS). Technology Dr. Humphries developed a computational method using convolutional neural network (CNN) algorithms to perform automatic, volumetric segmentation of the paranasal sinuses on CT. This allows efficient, reliable and objective measurement of sinus cavity volumes and opacification, which is an indication of mucosal thickening. Potential Applications Automated segmentation of the paranasal sinuses on CT using a CNN approach provides truly objective, volumetric quantitation of sinonasal inflammation, which can be used for diagnosis and stratification of disease severity. On sequential CT scans objective quantitation makes it possibly to measure changes precisely which can allow assessment of treatment efficacy. Potential applications include therapeutic trials and clinical assessment of patients with CRS. State of Development A functional prototype has been developed, and validated on over 2000 CT scans for analysis in multiple research studies and clinical trials. Publications Humphries SM, et al. Volumetric assessment of paranasal sinus opacification on computed tomography can be automated using a convolutional neural network. Int Forum Allergy Rhinol. 2020 Nov;10 (11):1218-1225. doi: 10.1002/alr.22588. PMID: 32306522 Beswick, DM. Machine learning evaluates improvement in sinus computed tomography opacification with CFTR modulator therapy. Int Forum Allergy Rhinol. 2021 May;11 (5):953-954. doi: 10.1002/alr.22722. PMID: 33140564 Beswick DM, et al. Impact of CFTR Therapy on Chronic Rhinosinusitis and Health Status: Deep Learning CT Analysis and Patient Reported Outcomes. Annals of the American Thoracic Society. 2021 Aug 26(ja). PMID: 34436985 Patent Status Patent pending. Inventors Stephen Humphries, PhD Licensing Status This technology is available for license. For Further Information, Contact: Emmanuel Hilaire, PhD Director Technology Transfer Office National Jewish Health 1400 Jackson Street, Room M206b Denver, CO 80206 Voice: 303.398.1262 HilaireE@njhealth.org